CN112149543B - Building dust recognition system and method based on computer vision - Google Patents
Building dust recognition system and method based on computer vision Download PDFInfo
- Publication number
- CN112149543B CN112149543B CN202010973045.2A CN202010973045A CN112149543B CN 112149543 B CN112149543 B CN 112149543B CN 202010973045 A CN202010973045 A CN 202010973045A CN 112149543 B CN112149543 B CN 112149543B
- Authority
- CN
- China
- Prior art keywords
- image
- building
- dust
- building dust
- pixel
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000000428 dust Substances 0.000 title claims abstract description 62
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000012544 monitoring process Methods 0.000 claims abstract description 21
- 238000012545 processing Methods 0.000 claims abstract description 20
- 230000000877 morphologic effect Effects 0.000 claims abstract description 8
- 230000007797 corrosion Effects 0.000 claims abstract description 6
- 238000005260 corrosion Methods 0.000 claims abstract description 6
- 238000001914 filtration Methods 0.000 claims description 11
- 238000010276 construction Methods 0.000 claims description 10
- 238000009499 grossing Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 claims description 3
- 230000003203 everyday effect Effects 0.000 claims description 2
- 238000010586 diagram Methods 0.000 claims 2
- FGUUSXIOTUKUDN-IBGZPJMESA-N C1(=CC=CC=C1)N1C2=C(NC([C@H](C1)NC=1OC(=NN=1)C1=CC=CC=C1)=O)C=CC=C2 Chemical compound C1(=CC=CC=C1)N1C2=C(NC([C@H](C1)NC=1OC(=NN=1)C1=CC=CC=C1)=O)C=CC=C2 FGUUSXIOTUKUDN-IBGZPJMESA-N 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 5
- 238000006243 chemical reaction Methods 0.000 abstract description 4
- 238000004891 communication Methods 0.000 abstract description 4
- 230000007547 defect Effects 0.000 abstract description 4
- 238000005259 measurement Methods 0.000 abstract description 4
- 230000009286 beneficial effect Effects 0.000 abstract 1
- 238000004364 calculation method Methods 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000009435 building construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 239000011800 void material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
Abstract
The invention discloses a building dust recognition system and method based on computer vision. And performing color model conversion on the processed image. The binary image morphological opening operation is performed, corrosion is performed first, then expansion is performed, the influence of fine objects is eliminated, the edge area is enlarged, the background and foreground images are obtained, and the background and foreground images are combined to form a mask image. And calculating the ratio of the pixel area of the extracted building dust to the total pixel area of the image by using a pixel area method, and judging whether the image has the building dust. The image has a stored image of building dust and outputs an identification alarm signal. The invention is a system for automatically completing the automatic identification and alarm of building dust by combining computer vision, image processing and modern communication technology, solves the defects of long measurement period, poor precision and the like of the existing dust detector, and is beneficial to the monitoring work of management staff.
Description
Technical Field
The patent relates to computer vision and image processing technology, in particular to an automatic recognition system for building dust by using a filtering process, image model conversion, channel separation, global threshold binarization, morphological corrosion and expansion, a watershed algorithm, dust recognition and an alarm signal output method.
Background
Dust particles are generated in each stage of building construction, so that real-time continuous monitoring of dust is needed. At present, the domestic dust monitoring method mainly comprises a dust detector taking an optical sensor as a counter, and has the advantages of large measurement error, long period and poor stability, and many monitoring methods can not timely give measurement results. Along with the rapid development of image processing and computer vision, the invention provides a building dust recognition method combining computer vision, image processing and modern communication technology, which can detect building dust targets in time and rapidly alarm, and overcomes the defects of building dust detection in the prior art. The wireless cameras with different angles are installed on a construction site, single-frame monitoring images are acquired by the cameras with different angles according to fixed time of each day, and the single-frame monitoring images are transmitted to an image information base through a wireless network. The method comprises the steps of calling an image of an information base, running VS2010 software to process the image, and judging whether a large-area dust pollution exists in a construction site or not by utilizing computer vision. If dust pollution exists, giving an alarm prompt, and reminding a manager of the construction site of carrying out dust fall measures. The method is based on a dust pollution recognition and recognition signal output mechanism of the monitoring terminal, is simple to operate, liberates a large amount of manual labor, and effectively reduces the maintenance cost of the instrument.
Disclosure of Invention
The invention provides a building dust recognition method by combining computer vision, image processing and modern communication technology. The defect that the existing building dust monitoring is time-consuming and labor-consuming and the measurement error is large is overcome. The method is simple to operate, economical and efficient, can sensitively and rapidly obtain the image information of the construction region, judges whether building dust is generated or not, gives out alarm identification signals, and brings great convenience to personnel monitoring the building dust.
In order to achieve the purpose, the technical scheme adopted by the invention is a building dust recognition scheme based on computer vision:
the building dust recognition system based on computer vision mainly comprises a building site image acquisition module, a building site image information storage module, a building site image processing module, a display for displaying building site image information and an alarm module. The building site image acquisition module is connected with the building site image information storage module, and the building site image information storage module is connected with the building site image processing module; the building site image processing module is respectively connected with the display and the alarm module; the image acquisition is carried out by acquiring images of surrounding environment of the building site at fixed time every day by wireless network cameras arranged at all angles of the building site, transmitting the images to an information storage module through a wireless network, storing image information, filtering and algorithmic analysis by an image processing module, and sending an alarm signal when the input image information reaches an alarm threshold value. Timely and efficient identification information is provided for monitoring personnel, and identification alarm signals are output in real time, so that the operation and maintenance cost of monitoring building dust is greatly reduced.
The monitoring, identification and alarm signal output of building dust in building site image is mainly characterized by that a series of building site images are undergone the process of image processing to obtain identification target, the target is undergone the process of calculation of pixel point proportion, and finally the output alarm signal is judged.
The steps are as follows:
the method comprises the steps of a), transmitting a single frame image of a construction site acquired by an image acquisition module to an image information storage module, and calling the single frame image of the image information storage module to perform image initialization processing, wherein the specific method is to adjust the size of the single frame image and the resolution of the image, and the formula is as follows:
pc=src.cols/dest.cols
pr=src.rows/dest.rows
scr.cols is the height of the original image, scr.rows is the width of the original image;
the dest.cols is the height of the adjusted image, and the dest.rows is the width of the adjusted image;
pc is the height scale, pr is the width scale;
the original image width and height are x×pc and y×pr respectively, and the output image width and height are x and y respectively, so that the newly adjusted image is not lost.
Step b) carrying out Gaussian filtering on the image with the adjusted size and resolution to retain the target characteristics of the image, and removing a large amount of noise pollution to the image in the processes of forming, transmitting and storing. The Gaussian filtering is a smoothing filter which carries out convolution operation on pixel points of an input image and a convolution module of a Gaussian kernel, carries out weighted average on the whole image and forms a filtered image array by one block of results. The weighting value is determined by the shape of a gaussian function, and the formula of the two-dimensional gaussian function is:
g (x, y) is the pixel value of the output image at the (x, y) point, and the distribution parameter sigma is the width of the filter;
x is the abscissa value of the pixel point, and y is the ordinate value of the pixel point;
the calculation formula is as follows by using the Gaussian kernel of 3×3:
g(x,y)={f(x-1,y-1)+f(x-1,y+1)+f(x+1,y-1)+f(x+1,y+1)+[f(x-1,y)+f(x,y-1)+f(x+1,y)+f(x,y+1)]*2+f(x,y)*4}/16;
g (x, y): outputting pixel values of the image at (x, y) points;
f (x, y): pixel values of the input image at (x, y) points;
and c) converting the image of the RGB color model after the two-dimensional Gaussian filtering into an image of an HSV color model, wherein the building dust feature in the image is more prominent after the image with building dust is converted into the HSV color model. The HSV color model is a color space of a hexagonal cone model, and parameters of the model include hue, saturation and brightness. The two model conversion relation calculation formulas are as follows:
V=max
max: representing the maximum value among the three channels of RGB;
min: representing the minimum in the three channels of RGB;
after the image in the step d) is converted into an HSV color model, the image of the HSV model is separated into images of three channels of hue, saturation and brightness. Finding out an optimal threshold interval according to the pixel value and the frequency distribution histogram of each channel image, adopting global threshold interval binarization, setting the pixel value of the pixel points of the traversing image in the threshold interval to be 255, and setting the pixel value not in the threshold period to be 0.
Step e), combining the binarized images of the three channels, and performing morphological opening operation on the newly generated images. Invoking the opencv functions erode and dialate erodes before expanding, and the open operation has the effect of eliminating fine objects, separating objects at the fine and smoothing the boundaries of larger objects.
The step of the corrosion algorithm:
(1) Scanning each element of the image with 3x3 structural elements
(2) AND operation with structural elements and binary images covered thereby
(3) If both are 1, the pixel of the resulting image is 1. Otherwise is 0
(4) Results: eliminating tiny meaningless noise points
The step of the expansion algorithm:
(1) Scanning each element of the image with 3x3 structural elements
(2) AND operation with structural elements and binary images covered thereby
(3) If both are 0, the pixel of the resulting image is 0. Otherwise is 1
(4) Results: filling the background void phenomenon.
And combining two foreground and background binarized images obtained by morphological opening operation into a 32-bit mask image, and dividing the target area by using the mask image as a mark.
And f) taking the combined mask image after corrosion and expansion treatment as a mark image of a watershed algorithm. The watershed algorithm is a mathematical morphology bottom-up recursion process based on topology theory, and is based on similarity between adjacent pixels as an important reference, so that pixels which are similar in spatial position and similar in gray value are connected with each other to form a closed contour so as to divide a dust object. The recursive formula is as follows:
representing the smallest pixel point of the gray value in the image I;
h min representing the minimum gray value in the image, h max The value with the maximum gray value;
X h+1 representing all pixel points, min on the gray value h+1 h+1 The minimum point of the gray value in the region is represented,is X h The point is in the region with gray value h and X h ∩X h+1 Representing the point where the two intersect, at the same region;
dividing the building dust-raising area by using a watershed algorithm, counting the number of the pixels of the dust-raising target area and the number of the pixels of the whole image by using a pixel area method, and calculating the ratio value of the number of the pixels of the dust-raising target area and the number of the pixels of the image. The formula is as follows:
scale_img=whitecount/pixekcount
while the whistecount is the number of pixels in the building dust-raising target area, the pixekcount is the number of total pixels in the whole image, and scale_img is the ratio value of the two;
and h) calculating a proportion value of the building dust area and the whole image, marking the image with the proportion value larger than a threshold value with a problematic label, carrying out target recognition on the building dust, and giving a target recognition alarm signal. The program stores the problematic photo, and is convenient for monitoring personnel to check information.
The invention aims to solve the defects of long detection period, poor precision and high maintenance cost of the existing building dust. A building dust recognition method combining computer vision, image processing and modern communication technology is provided. The method is economical and efficient, can sensitively and rapidly acquire the image information of the construction region, judges whether the construction dust exists or not, gives out alarm identification signals, and brings great convenience to personnel monitoring the construction dust.
Drawings
Fig. 1 is a flowchart of an algorithm.
Fig. 2 is a view of building dust target monitoring. a is an original drawing; b is a dust identification chart.
Fig. 3 is a graph of real-time monitoring signal output.
Detailed Description
Fig. 1 is a flow chart of overall design of building dust monitoring, and the core part is an image processing algorithm process. The pictures with building dust are subjected to target monitoring and output monitoring identification signals, and the experiments can achieve good effects. The specific implementation mode is as follows:
1. the single frame image acquired by the information base is initialized in size and resolution, the size of the image is set to 400 x 400, and the image is named as 'g_srchimage'. The initialized image is subjected to gaussian filtering processing, so that σ=0.7 in the gaussian function.
2. The preprocessed RGB image is converted into HSV model pictures, the pictures after model conversion are separated into three channels of pictures, and global binarization operation is carried out respectively. Three binarized pictures himg, simg, vimg are obtained and combined into one binarized picture binary_image.
3. Morphological opening operation is carried out on the binary_image, the opencv function erode corrosion operation is called for 6 times to obtain a foreground picture, and the function dialate expansion operation is called for 6 times to obtain a background picture. The foreground picture and the background picture form a mask picture marker, and the mask picture is used as a mark to extract a target area by using a watershed algorithm.
4. And calculating the proportion area of the dust raising area by adopting a pixel area method, marking a problem label when the proportion is larger than 0.06, storing a picture with building dust raising and outputting an alarm identification signal.
Claims (8)
1. The building dust recognition method based on computer vision comprises a building image acquisition module, a building image information storage module, a building image processing module, a display and an alarm module;
the building site image acquisition module is connected with the building site image information storage module, and the building site image information storage module is connected with the building site image processing module; the building site image processing module is respectively connected with the display and the alarm module;
in the building site image acquisition module, wireless network cameras arranged at all angles of a building site acquire images of surrounding environments of the building site at fixed moments every day; transmitting the image information to a building site image information storage module via a wireless network to store the image information; filtering and algorithmically analyzing by a building image processing module, and sending an alarm signal when the input building image information reaches an alarm threshold value; displaying building image information by a display;
the monitoring, identification and alarm signal output of building dust in building site images is mainly realized by performing image processing on a series of building site images to obtain an identification target, calculating the pixel point proportion of the target and finally judging the output alarm signal;
the method is characterized in that: the method is implemented as follows,
the method comprises the steps that a) an image acquisition module acquires a single frame image of a construction site, transmits the single frame image to an image information storage module again, calls the single frame image of the image information storage module to perform image initialization processing, and adjusts the size of the single frame image and the resolution of the image;
step b), performing Gaussian filtering on the image with the adjusted size and resolution to retain target characteristics of the image, and removing a large amount of noise pollution on the image in the forming, transmitting and storing processes; the Gaussian filtering is a smoothing filter for carrying out convolution operation on pixel points of an input image and a convolution module of a Gaussian kernel, carrying out weighted average on the whole image and forming a filtered image array by the results;
step c), converting the image of the RGB color model after the two-dimensional Gaussian filtering into an image of an HSV color model, and after converting the image with building dust into the HSV color model, enabling the building dust characteristic in the image to be more prominent; the HSV color model is a color space of a hexagonal cone model, and parameters of the model include hue, saturation and brightness;
after the image in the step d) is converted into an HSV color model, separating the image of the HSV model into images of three channels of hue, saturation and brightness; finding out an optimal threshold interval according to the pixel value and the frequency distribution histogram of each channel image, adopting global threshold interval binarization, setting the pixel value of the pixel points of the traversal image in the threshold interval to be 255, and setting the pixel value not in the threshold period to be 0;
step e), combining the binarized images of the three channels, and performing morphological opening operation on the newly generated images; the open operation has the effects of eliminating fine objects, separating objects at the fine and smoothing the boundary of larger objects;
step f) the mask images combined after corrosion and expansion treatment are used as mark images of a watershed algorithm;
dividing a building dust-raising area by a watershed algorithm, counting the number of pixels in a dust-raising target area and the number of pixels in the whole image by adopting a pixel area method, and calculating the ratio value of the number of pixels in the dust-raising target area to the number of pixels in the image;
step h), calculating a proportion value of a building dust area and the whole image, marking the image with the proportion value larger than a threshold value with a problematic label, carrying out target recognition on building dust, and giving a target recognition alarm signal; the program stores the problematic photo, and is convenient for monitoring personnel to check information.
2. The building dust identification method based on computer vision according to claim 1, wherein the method comprises the following steps: and transmitting the single frame image of the construction site to an information base by using a camera, and adjusting the size of the image and the size of the resolution.
3. The building dust identification method based on computer vision according to claim 1, wherein the method comprises the following steps: noise pollution received in the image forming and transmitting process is filtered by Gaussian filtering, and isolated pixel points and pixel blocks are eliminated under the condition that the detail characteristics of the image are maintained.
4. The building dust identification method based on computer vision according to claim 1, wherein the method comprises the following steps: converting an RGB color model of an image into an HSV color model, and separating the image of the HSV color model into three channels of brightness, saturation and brightness; the global threshold of each channel image is binarized, and three binarized images are combined into one binarized image.
5. The building dust identification method based on computer vision according to claim 1, wherein the method comprises the following steps: and performing morphological opening operation on the binarized image, eliminating fine objects, corroding and then expanding the fine objects to obtain foreground and background images, and merging to obtain a mask image.
6. The building dust identification method based on computer vision according to claim 1, wherein the method comprises the following steps: according to a mask diagram obtained by morphological opening operation, the mask diagram is used as a threshold value mark image of a building dust recognition watershed algorithm; the watershed algorithm is that the threshold value marked image sequences the pixel gray level of the building dust-raising target area from low to high, then inundation is realized from the pixel value to high, the boundary point of the highest gray value of the marked image is the watershed, and the boundary point is the edge information of the building dust-raising target area; and (5) segmenting and extracting the dust of the target building by using a threshold marked watershed algorithm.
7. The building dust identification method based on computer vision according to claim 1, wherein the method comprises the following steps: and judging whether to output an alarm identification signal by using a pixel area method according to the ratio of the sum of pixel areas for extracting building dust to the total pixel area of the image.
8. The building dust identification method based on computer vision according to claim 1, wherein the method comprises the following steps: judging that building dust exists in the image, storing the image and outputting an identification alarm signal of the building dust.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010973045.2A CN112149543B (en) | 2020-09-16 | 2020-09-16 | Building dust recognition system and method based on computer vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010973045.2A CN112149543B (en) | 2020-09-16 | 2020-09-16 | Building dust recognition system and method based on computer vision |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112149543A CN112149543A (en) | 2020-12-29 |
CN112149543B true CN112149543B (en) | 2024-02-27 |
Family
ID=73893067
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010973045.2A Active CN112149543B (en) | 2020-09-16 | 2020-09-16 | Building dust recognition system and method based on computer vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112149543B (en) |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112785710B (en) * | 2021-01-28 | 2021-12-21 | 湖北省国土测绘院 | Rapid unitization method, system, memory and equipment for OSGB three-dimensional model building |
CN113888397A (en) * | 2021-10-08 | 2022-01-04 | 云南省烟草公司昆明市公司 | Tobacco pond cleaning and plant counting method based on unmanned aerial vehicle remote sensing and image processing technology |
CN114782415A (en) * | 2022-06-16 | 2022-07-22 | 长春融成智能设备制造股份有限公司 | Filling barrel surface abnormal state real-time monitoring method based on machine vision |
CN114842349B (en) * | 2022-07-01 | 2022-09-06 | 山东高速德建建筑科技股份有限公司 | Building construction environment protection method and system based on information technology |
CN117036250B (en) * | 2023-07-14 | 2023-12-26 | 小鲲智能技术(广州)有限公司 | Method and device for judging floc sedimentation performance based on visual algorithm |
CN117274884B (en) * | 2023-11-21 | 2024-02-20 | 赣江新区慧工科技有限公司 | Construction dust pollution event detection method and system based on image recognition |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106203265A (en) * | 2016-06-28 | 2016-12-07 | 江苏大学 | A kind of Construction Fugitive Dust Pollution based on unmanned plane collection image is derived from dynamic monitoring and coverage prognoses system and method |
CN107578045A (en) * | 2017-09-19 | 2018-01-12 | 北京工业大学 | A kind of Underwater targets recognition based on machine vision |
CN109460705A (en) * | 2018-09-26 | 2019-03-12 | 北京工业大学 | Oil pipeline monitoring method based on machine vision |
-
2020
- 2020-09-16 CN CN202010973045.2A patent/CN112149543B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106203265A (en) * | 2016-06-28 | 2016-12-07 | 江苏大学 | A kind of Construction Fugitive Dust Pollution based on unmanned plane collection image is derived from dynamic monitoring and coverage prognoses system and method |
CN107578045A (en) * | 2017-09-19 | 2018-01-12 | 北京工业大学 | A kind of Underwater targets recognition based on machine vision |
CN109460705A (en) * | 2018-09-26 | 2019-03-12 | 北京工业大学 | Oil pipeline monitoring method based on machine vision |
Non-Patent Citations (1)
Title |
---|
基于计算机图像处理的智能监控技术研究;李倩;;安阳师范学院学报(第05期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN112149543A (en) | 2020-12-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112149543B (en) | Building dust recognition system and method based on computer vision | |
CN106651872B (en) | Pavement crack identification method and system based on Prewitt operator | |
TWI409718B (en) | Method of locating license plate of moving vehicle | |
CN110322522B (en) | Vehicle color recognition method based on target recognition area interception | |
CN108181316B (en) | Bamboo strip defect detection method based on machine vision | |
CN106682665B (en) | Seven-segment type digital display instrument number identification method based on computer vision | |
CN113887412B (en) | Detection method, detection terminal, monitoring system and storage medium for pollution emission | |
CN109087363B (en) | HSV color space-based sewage discharge detection method | |
CN110415208A (en) | A kind of adaptive targets detection method and its device, equipment, storage medium | |
CN110175556B (en) | Remote sensing image cloud detection method based on Sobel operator | |
CN113706566B (en) | Edge detection-based perfuming and spraying performance detection method | |
CN114219773B (en) | Pre-screening and calibrating method for bridge crack detection data set | |
CN115731493A (en) | Rainfall micro physical characteristic parameter extraction and analysis method based on video image recognition | |
CN112861654A (en) | Famous tea picking point position information acquisition method based on machine vision | |
CN114331986A (en) | Dam crack identification and measurement method based on unmanned aerial vehicle vision | |
CN111665199A (en) | Wire and cable color detection and identification method based on machine vision | |
CN111768455A (en) | Image-based wood region and dominant color extraction method | |
CN111461076A (en) | Smoke detection method and smoke detection system combining frame difference method and neural network | |
CN112634179B (en) | Camera shake prevention power transformation equipment image change detection method and system | |
CN113971681A (en) | Edge detection method for belt conveyor in complex environment | |
CN105787955A (en) | Sparse segmentation method and device of strip steel defect | |
CN113052234A (en) | Jade classification method based on image features and deep learning technology | |
CN108830834B (en) | Automatic extraction method for video defect information of cable climbing robot | |
CN114758139B (en) | Method for detecting accumulated water in foundation pit | |
CN112488031A (en) | Safety helmet detection method based on color segmentation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |